Now that 64-slice scanners have come on the market, coronary CT angiography (CTA) has a lot of doctors enthusing. They describe what the scanners can do as a watershed technology that will change cardiac imaging forever. Radiologists see opportunities, but so do cardiologists. More than radiologists, cardiologists face changes in their practice patterns.

For the first time, CT scanners can consistently image lesions of the native coronary arteries. This is a breakthrough that radiologists and cardiologists have been waiting for. Oddly, both groups of specialists face learning curves for CT angiography, but both are rushing to get involved. Radiologists want their specialty to dominate the new cardiac CT imaging in the way it has dominated other CT imaging up to now. Cardiologists are mounting their own effort to control the cardiac CT imaging landscape. Since the cardiologists have the patients, they have an advantage. But radiologists know the chest, not just the heart, and this gives them advantages too. At least they hope so.


Doctors use words like “promised land,” “holy grail,” and “gold standard” to describe the coronary arterial imaging possible with the latest generation of CT scanners.

J. Bayne Selby, Jr, MD, FSCVIR, is professor of radiology in the division of vascular and interventional radiology at the Medical University of South Carolina (MUSC), Charleston. He has watched cardiac CT imaging evolve.

“If you go back to the two slice, we started calcium scoring; with the quad we could see coronary artery bypass grafts; when we got to 16, we added congenital heart disease, pulmonary veins, and anomalous coronary arteries. We added these at a level that we felt we could do the diagnosis,” Selby says. “At 64 slices we have added the native coronary arteries and stents. The jump to 64 is really opening up the coronary arteries. Everybody is focused on the coronary arteries. That’s where the money is.”

Osman Ratib, MD, PhD, FAHA, is professor and vice chair of information systems in the radiology department at the University of California at Los Angeles (UCLA). Ratib is boarded in both radiology and cardiology. He says he was so excited when UCLA got a 64-slice CT scanner that he imaged his own heart just to see what the machine could do.

“I did 3,700 images in one run,” he says. “It was at the highest temporal resolution of the whole heart…. I don’t want to oversell it, but the 64-slice is an amazing improvement in quality.”

Asked about the examination outcome, he says, “Oh, I have great coronaries.”

Ratib says that, when he did coronary CT angiograms with the school’s old 16-slice scanner, he was able to conclusively identify the presence or absence of arterial stenosis or other anomalies in only 30% to 40% of the patients he scanned. With the 64, he says that the identification rate is much higher. In the 60 to 70 studies he has done so far, he says, only two have been inconclusive. “With the 64, we are seeing the plaques better and better.”

The 64-slice CT still cannot differentiate types of plaque as well in the coronary arteries as it can in the carotids, Ratib says, because of the smaller size of the heart vessels and the heart motion. “There is some indication we may be able to differentiate plaque in the carotids, and they are as important as the coronaries,” he says.

What the new CT scanners can do with apparent near perfection is rule out heart disease in questionable or asymptomatic patients complaining of chest pain. This is one area where the machines are expected to save patients from cardiac catheterizations when no stenting or angioplasty is called for. Hospitals, which will not have to admit these patients, will also save money.

“CT is excellent for ruling out disease,” says Norbert Wilke, MD, who like Ratib is double boarded in both radiology and cardiology. “If you have a negative calcium score and a negative coronary angiogram then you have a 95% chance of not having heart disease. This is the number one indication to rule out disease.”

Wilke, who is associate professor of medicine and chief of cardiovascular MR and CT at the University of Florida Health Science Center in Jacksonville, does add a caveat about the new CT’s ability, however. “If you do have heart disease,” he says, “the CT is not so good at grading stenosis. So then you’ve got to go to MR or the cath lab. Once you have excluded disease, the CT is a very good test to rely on.”

The image quality on the new CT scanners is said to be so good that it is akin to looking at a high-resolution arterial photograph. Carter Newton, MD, FACC, is a cardiologist and imaging consultant now living in Tucson after decades of practicing in Los Angeles. Newton compares the new CT imaging with what digital cameras can do in the photographic world. The level of detail is outstanding, he says. “In the same way that digital cameras have replaced film cameras, this CT imaging is going to replace invasive angiography,” he says. “It’s less ordeal for the patient and an equivalent amount of radiation. Doctors are going to accept this exam in higher and higher numbers.” Newton is so enthusiastic about the new CT that he is planning a training site in Arizona for doctors to fly in and learn the imaging techniquesfrom him.

The CT coronary artery examinations are not just single images. They are hundreds or thousands of slices that compose an imaging run, and they can be manipulated on the workstation in a number of ways.

J. Bayne Selby, Jr, at MUSC says that postprocessing of the CT examinations is “almost the bigger part of the game now.” He says the sheer volume of images is daunting enough, not to mention the way they can be manipulated at the workstation by various viewing techniques in two and, more important, three dimensions.

CTA Applications

Aside from detecting or eliminating coronary artery disease in asymptomatic patients, Michael Shen, MD, FACC, head of cardiac imaging at Cleveland Clinic, Weston, Fla, identified the following additional CT applications.

  • Evaluating patients with equivocal results from stress tests. “In this population, we can do the CTA and we won’t have to do a cath, which is invasive,” he says.
  • Evaluating patients with high risk factors like high cholesterol and diabetes who are asymptomatic. For these patients plaque can be identified early and treatment started to prevent heart attacks.
  • Evaluating ventricular function. The CT can image the heart in both the systolic and diastolic phases and can measure ventricular volumes or ejection fractions more accurately than nuclear stress testing can, Shen says.
  • Checking other heart chambers, especially the left atrium and the pulmonary veins before and after vein ablations. The CT images can indicate how many veins may need ablation. Afterwards, ablated veins can be checked for thrombosis that may need to be stented, says Shen.
  • Presurgical evaluations of the heart. Especially with patients who have had bypass surgery and need follow-up, the surgeon will need to know where the grafts are, Shen says. “The surgeon does not want a surprise, that’s another big application.”
  • Finally, there is CT’s ability to image the heart as part of a larger chest scan, Shen notes, so the coronary CTA doubles as a generalized chest scan in which tumors, aneurysms, embolisms, and other anomalies may be identified. This is where radiologists undoubtedly stand on firmer interpretive ground than cardiologists do.

—G. Wiley

“When we reconstruct at 0.75 mm, it will give you 965 images, and that’s just in the axial plane, not the sagittal or coronal,” he says. He says using 3-D techniques can save time by pinpointing “where you want to zone in and look more closely.” Even so, he adds, the data from the scan of one patient is so voluminous that to look at it in all the ways it can be manipulated is a practical impossibility. There is so much data that all of it cannot be archived in most institutions.

“If you tried to save all the raw data, it would bring the system to its knees,” Selby says. “Places that are doing research like we are will need huge optical drives and fast transfer rates. Most good hospitals will do a number of reconstructions and put those on the PACS (picture archiving and communications system) and save those, but they won’t be able to save every bit. What they save will be manipulable in some ways, but not in every possible way.”

U. Joseph Schoepf, MD, is an associate professor of radiology at MUSC and a colleague of Selby’s. Schoepf says the new 64-slice CT is so good at detecting coronary artery disease that it is replacing the more traditional CT test for calcium scoring.

“We have four or five CT angiograms per day but only about one calcium scoring per day with our scanner,” he says, “which tells you how much coronary CTA is being accepted. It has rapidly replaced calcium scoring as a much more valuable test.”


With cardiac CT testing gaining ground quickly, doctors and health systems are struggling to define the clinical applications for the examinations. The techniques are too new to have accumulated large repositories of clinical data.

Michael Shen, MD, FACC, is head of cardiac imaging at Cleveland Clinic Florida in Weston. Shen says where coronary CTA will make the most difference is in identifying those patients with plaque buildup going outward from the lumen, which he calls “positive remodeling.” It is the positive remodeling group of patients whose plaque cannot be detected with stress testing or cardiac catheterization, for whom coronary CT angiography holds the most promise, Shen says. “That is the key for the CTA.” It is particularly key because this group of patients, Shen says, accounts for about 70% of all heart attacks. These patients, with no prior heart disease history, often die because a fatal heart attack is the first sign of trouble for them, he says.

But Shen breaks the CT applications down further than just saving lives in the positive remodeling group (see sidebar CTA Applications).

Above all other utilizations, however, is that of quickly and efficiently detecting or eliminating coronary artery disease in previously undiagnosed patients, particularly those with high risk factors and family histories that would make them suspect.

Tim Attebery is CEO of the South Carolina Heart Center, headquartered in Columbia. The center has 28 cardiologists working out of a half-dozen locations. It is heavily into coronary CTA.

“I would make the claim that we’re at the frontier of this movement in the United States,” says Attebery. “We probably do as many or more cardiac CTAs than anyone. We are the only cardiology group with a 64-slice CT.”

Attebery ranks the impact of coronary CTA alongside that of the stent in revolutionizing heart care.

“There are probably 60 to 80 million Americans who have never seen a cardiologist who have subclinical disease and who are at risk,” he says. “For 35% to 40% of them, the condition they present with will be death.”

It is this huge asymptomatic pool of potential heart attack victims that Attebery and others see coronary CTA as possibly saving. “I believe the 64-slice CT is going to make a significant difference,” Attebery says.

(Click the image for a larger version.)
(Click the image for a larger version.)

Figure. 52-year-old-man presenting to the emergency department with acute chest pain. He had a smoking history of 25 pack years. Contrast enhanced 64-slice MDCT angiography of the entire thorax with a gantry rotation time of 330 ms and retrospective ECG gating rules out acute pulmonary embolism as a reason for chest pain (a). Focused reconstruction of the coronary arteries shows diffuse atherosclerotic disease (b). Maximum intensity projection in a right anterior oblique perspective of the left anterior descending (LAD) coronary artery shows atherosclerotic plaque in the inferior, antero-medial wall of the LAD (c). The plaque is mildly obstructive and consists of noncalcified tissue adjacent to a calcified nodule. Analysis of lung-window reconstructions of the entire chest reveals incidental squamous cell carcinoma (d) of the left upper lobe of the lung. Image courtesy of U. Joseph Schoepf, MD, Medical University of South Carolina, Charleston.

(Click the image for a larger version.)
(Click the image for a larger version.)


Before coronary CTA can have its impact, however, it has to be far more widely adopted. Currently, only three states—New York, New Jersey, and South Carolina—have Medicare CPT codes in place for the cardiac CT procedure. In all others, it must be approved as a chest scan, a radiological code. Private payors are asking for clinical documentation or a professional stamp of approval from subspecialist groups before they add coronary CTA to their approved lists. Both radiologists and cardiologists are working to stake out the coronary CTA turf as this adoption process proceeds.

UCLA’s Osman Ratib says that there are “a lot of sort of semi-secret meetings to talk strategy” going on in the medical community to determine how the rollout of coronary CTA will take place. Not all of the strategizing involves the so-called turf war between radiologists and cardiologists either.

The University of Florida’s Norbert Wilke says the turf war is not even the big story in the rollout of coronary CTA. The real story, he says, is how the CT examinations are going to alter the playing field for cardiologists themselves by superseding traditional examinations like diagnostic cardiac catheterization, echocardiography, and especially nuclear stress testing.

According to the South Carolina Heart Center’s web site ( ), more than 5 million nuclear stress tests (myocardial perfusion scans) are done in the United States every year. Wilke says these scans form the income backbone of many cardiology groups. If coronary CTA supersedes them, as Wilke predicts, then these cardiologists will have to find ways to make up the income. Cardiac CTA is one way, except they are a much less profitable procedure than the hours-long nuclear stress tests.

Playing with Fire

Alan Kaye, MD, FACR, is president of Advanced Radiology Consultants (ARC) in Bridgeport, Conn. The group of 17 radiologists reads for the 450-bed Bridgeport Hospital and at eight imaging centers the group owns in southern Connecticut, Kaye says.

ARC is about to install a 32-slice CT scanner and begin offering outpatient coronary CTA examinations. Kaye says he expects cardiologists to be part of the referring physician pool that sends cardiac CT patients to ARC. The reason is simple. Connecticut is a certificate of need (CON) state, and ARC has gone through the lengthy procedure to get approval for the scanner from state CON officials.

Because the CON approval is lengthy, Kaye says cardiologists will not be able to purchase and deploy their own CT scanners without going through the CON process.

“We want the opportunity to demonstrate to the cardiologists that we can work together and provide them with information that will help them take care of their patients,” Kaye says. He says one benefit of having radiologists do the cardiac CTAs is that other specialists, like cardiologists, will also see the scans. “Cardiologists can get away with something that is not optimal because nobody sees their scans since they self-refer,” he says.

Kaye says he is aware that cardiologists will be “chomping at the bit to get their own scanners.” But he knows the CON impediments will hold them back for a while. ARC’s scanner is expensive, more than $1 million, Kaye says. But the group is gambling that cardiologists will send them the patients needed to make the machine profitable. If not, says Kaye, the scanner can always be used for all other CT purposes.

“Perhaps this is idealistic,” he says, “but I think it deserves a shot. I know I’m playing with fire…”

—G. Wiley

One way to ease the pain for nuclear scan cardiologists is to increase the reimbursement rates for cardiac CTA, Wilke believes. “The payors are very confused,” Wilke says. “What do you think the nuclear people are going to tell them? Here comes the CT community saying, We’ve got all these beautiful pictures.’ But the payors are saying we’re already bankrupt from paying for all the nuclear testing in this country.

“So we have to make a plan,” Wilke says, “where if you don’t do nuclear testing, you get compensated. We need to increase by 80% to 100% the financial incentives for CT and MRI. If the nuclear people aren’t going to compromise, we’re not going to get anywhere. It depends on how all this is lobbied, but that is going to be the discussion.”

Some people think Wilke is overstating the case, but he may not be.

On January 8, two separate societies were formed to promote and guide the adoption of coronary CTA. They are now going through a legal process to merge. Wilke is a board member of one of them, the Society of Cardiovascular Computed Tomography (SCVCT). The SCVCT counts radiologists among its members. The second group, the Society of Cardiac Computed Tomography (SCCT), appears to have more of a cardiology focus. A third group, the North American Society of Cardiac Imaging (NASCI), also comes into the picture. It is traditionally a radiology-oriented group without a focus on a specific modality.

Michael Poon, MD, FACC, is director of cardiovascular medicine and the integrated imaging program at Cabrini Medical Center, a 400-bed hospital in New York City. Poon is a board member and president of SCVCT. He says he was named president because of his role in pushing through CPT codes for cardiac CTA in New York and New Jersey.

At first, Poon plays down the potential impact of coronary CTA on nuclear stress scanning. He notes that one of the indications that must be met under the CPT codes for Medicare approval of a CTA examination is that the patient must first have an equivocal result from a noninvasive perfusion test, which for the most part means a nuclear stress test, he says. But then Poon says that patients are already coming to him who are willing to pay out of pocket for coronary CTA in order to avoid the time and expense of a nuclear stress scan.

“CT is a better test for coronary artery disease,” he states. “Unquestionably, CTA is a better test,” he says. “There’s no other test that shows you the vessels…. If the CT is negative, what’s the point in doing a nuclear stress test? It’s lengthy, and it doesn’t give you important information if your symptom is not classic.”

The CT angiogram is also less expensive, Poon says. He says nuclear stress tests cost $3,000 or more. The reimbursement from Medicare for cardiac CTA under the CPT codes he championed, Poon says, is between $586 and $764 depending on the exact procedure.

Poon notes, however, that nuclear stress tests have a long and well-researched history. “We know a lot about how to use the perfusion study,” he says, “but CTA is new and we need more study.” He also notes there are glitches with CTA that will keep nuclear testing alive. “When you have a lot of calcium,” he says, “CT is not very good. When the x-ray hits the calcium, it creates a sort of blooming effect and can make the lesion look worse than it is. So then you would need clarification with a nuclear stress test.”

Poon also notes that the advent of coronary CTA will probably cut down on the number of cardiac catheterizations done purely for diagnosis. “Thirty percent of catheterizations are negative,” he says. “That patient has been put at risk with one chance in 1,000 of a heart attack or a stroke.”

MUSC’s U. Joseph Schoepf says cardiologists will not protest the loss of these noninvasive catheterizations because the procedure is so poorly reimbursed that it makes little money. “So if we use CTA as an added test in those patients, the cardiologists can save [catheterization] room time and use the room for patients who need it [for stenting or angioplasty], which is much better reimbursed,” Schoepf says.

Poon says his strategy to get private payors to accept coronary CTA as reimbursable is to push for nationwide CPT codes from Medicare first. “Then we’ll go to the private payors,” he notes. “Compared to the cost of a nuclear stress test, it will be a no-brainer for them.”

Wilke raises another problem for CTA that may need clarification through the standards and protocols that the SCVCT and the SCCT hope to adopt and promote. That problem is radiation dosage. “It can be up to five millisieverts on the 64 slice,” he says, “and that’s very high. If you think about the scanner, you have to think about the tumors you may induce. We haven’t talked much about it.”


In the final analysis, one question remains unanswered: which group of subspecialists will establish itself as predominant in coronary CTA testing and interpretation? This is a question that greatly concerns radiologists. It does not appear to concern cardiologists in the same way. They see themselves as clearly in the driver’s seat, despite physician self-referral statutes that may present roadblocks.

MUSC’s J. Bayne Selby, Jr, a radiologist, says it is imperative that radiologists get trained in greater numbers to do coronary CTA. Otherwise, the procedure will be lost to cardiologists, he says.

Selby says that in the real world both cardiologists and radiologists will do coronary CTA. “Who will do it in any institution is who will have the interest,” he says. “It’s not a part-time job.”

But Selby says coronary CTA is a chance for radiologists to reassert themselves in the domain of heart imaging after decades of losing ground to cardiologists in imaging tests like nuclear stress scans and echocardiography. “I do think the radiology leadership is clearly seeing this as an obligation that we have to train our young people to be knowledgeable again about cardiac imaging. Our responsibility is to put those people out,” Selby says.

Selby says his big fear is that radiologists will give ground without a fight. “The subjective part, and I hate to say this, is that most radiologists who aren’t involved will say that cardiologists are going to take it anyway, so why bother. That’s what I hear. They say they still want the training, but they have a fatalistic attitude.”

Selby says radiologists have a big advantage since they are used to sitting at workstations manipulating images, and since they know the chest structures other than the heart. He says if cardiologists do the scans and the interpretations, they will do them “in a much thinner way” than radiologists would. “They’re not fully invested in how workstations work and what pitch means. They’ll just ask, is there plaque or not plaque? That’s harsh, but that’s how I see it being done.”

At MUSC, the radiologists read the cardiac CTA scans and then consult with the school’s cardiologists. Anathema to Selby would be to have radiologists doing overreads for nonheart anomalies in chest scans where cardiologists would do the cardiac interpretations themselves. “I hear that from a lot of radiologists. I think that’s the worst of all worlds,” he says. “There’s no reason to have two separate specialists.”

The trained radiologist’s advantage would be that he or she could do the entire chest study, including the heart, he says.

Selby says utilization rate studies for cardiologists who self-refer patients may help convince payors that radiologists can be more objective in delivering cardiac CTA. “This is one time where the payors really could do a good thing. I say to them, please look at this. You tell us what makes sense. But the payors just sort of spin around.”

Selby’s fellow radiologist at MUSC, U. Joseph Schoepf, says, “Radiology has a good model to integrate cardiac CT into a practice.” That is because a radiology group would use the scanner for all sorts of tests, not just cardiac ones. “A cardiologist would need to have a dedicated center or an extremely high group throughput to make cardiac CTA effective,” Schoepf adds. Unlike Selby, though, Schoepf is not daunted by the thought of radiologists overreading the nonheart parts of a chest scan. He says the over reads should be required.

“Everybody is getting geared up for a turf war, unfortunately,” he says, “which is not what we need. I’m not sure that we as radiologists need to fight that fiercely because part of it is going to fall to us anyhow, because that’s what we do.”

For their part, cardiologists appear to be laying the groundwork for a quick proliferation of 64-slice CT scanners into their practices. Carter Newton in Tucson is not the only one gearing up to teach cardiologists how to read coronary CTA. Ironically, MUSC is offering mini-courses, which cardiologists can attend.

“This technology is not that difficult,” Newton says. “I could teach you to read these scans in about a week. The scanners are getting that good that you’re not going to have to be a medical genius to learn this. They [radiologists] act like its landing a man on the moon, but it isn’t.”

Newton makes the case that cardiology is an image-based specialty. He calls it “an extension of the stethoscope.” He adds, “If you took imaging away from cardiologists, we’d be out selling our houses; that’s just a reality.”

South Carolina Heart Center’s Tim Attebery puts it bluntly. “The disease is managed by cardiologists,” he says. “When it’s time to figure out what to do, they ask the cardiologists. In the end the radiologist will always work with the cardiologist through these issues. There are 50 million heart patients in the United States. Radiologists are not going to have the ongoing responsibility of managing their disease, so there is no turf battle.”

As societies like SCVCT and SCCT gear up, it is clear that cardiologists want a big share of the cardiac CTA pie. Attebery says his center used to routinely have radiologists overread the nonheart portions of the cardiac CTA tests. But now, he says, that has been cut back. Overreads are only done when cardiologists think something is amiss in a chest scan. Attebery says his liability insurance carriers “don’t see any different risk exposure with this than any other area where we have to demonstrate competency.”

Like others, Attebery says the point may soon be moot anyway since CT vendors are developing machines that will focus down on the heart only, letting cardiologists off the hook for anomalies in the rest of the chest. They will not see the rest of the chest, so how could they know?

Of course, for now coronary CTA has to be billed as a chest scan in most states, but Attebery says Medicare CPT codes for the cardiac test alone will be universally in place by 2007 at the latest. Then radiologists will not be needed even for overreads unless at a cardiologist’s bidding, he says. Despite this, Attebery advocates that radiologists and cardiologists work together. “I prefer that we cooperate with the other societies. It’s better for the technology if the larger organizations can support it.”

No matter how the coronary CTA landscape finally takes shape, the effort to shape it is taking place at this very moment. Daniel S. Berman, MD, is director of cardiac imaging at the Cedars-Sinai Medical Center in Los Angeles. He is also the spokesperson for the SCCT.

Berman says standards and protocols and accreditation and certification procedures for coronary CTA reimbursement have to be quickly developed. “If radiologists circle the wagons on cardiovascular CT, it will be a severe setback for the field,” he says, “but I don’t think they can be successful in keeping cardiologists out. Cardiologists will get training. There will be methods to get cardiologists accredited, because cardiology will be the biggest growth market for scanner sales…. You have big purchasing power about to be unleashed in the area of cardiac CT…. For radiologists to think they can get around this with self-referral rules is a pipedream.”

George Wiley is a contributing writer for Decisions in Axis Imaging News.